Ming Sun, Martijn L. Manson, Elizabeth CM de Lange, Tingjie Guo
Division of Systems Pharmacology and Pharmacy, Leiden Academic Center for Drug Research, Leiden University
Introduction/Objectives: Viral encephalitis is a rare but life-threatening central nervous system (CNS) infection commonly caused by herpesviruses with a mortality rate up to 70% without antiviral therapies [1]. Acyclovir stands as the primary treatment yet its standard dosing was based on weaker evidence or from the 20th century [2-4], suggesting the necessity to revisit the current dosing strategies. Pharmacokinetic-pharmacodynamic (PK-PD) modeling approach could faciliate to develop optimal dosing. However, direct human brain data are scarce due to ethical concerns, complicating the conventional data-driven model development. Physiologically-based pharmacokinetic (PBPK) modeling, utilizing known drug properties and physiology, offers an alternative solution but is hampered by the use of fixed parameters and limited data. For this, the Bayesian method that incorporate prior knowledge and observed data can be applied to enhance model utility, addressing parameter uncertainty and sparse CNS data challenges [5]. The aim of the current study was to develop a Bayesian PBPK analysis framework to evaluate acyclovir treatment in viral encephalitis with the consideration of both drug efficacy and drug toxicity.
Methods: We developed the model to describe the acyclovir CNS pharmacokinetics (PK) based on a previously developed LeiCNS model framework including various brain compartments [6]. The model was developed using the Bayesian methodology that integrated the information from measured PK data (plasma and brain subarachnoid space (SAS) [7, 8]) and prior knowledge of the physiology parameters[6]. We then used the developed model to simulate acyclovir concentrations in brain extracellular fluid (brainECF) considering a standard dosing regimen (10 mg/kg TID) and two alternatives (15 mg/kg, 20 mg/kg TID). Therewith, we calculated the probability of target attainment (PTA) defined as the probability that acyclovir concentration remains above the IC50 of herpes simplex virus (HSV-1 and HSV-2) and varicella-zoster virus (VZV) for at least 12 hours in a 24-hour period to evaluate the efficacy of these dosing regimens. A range of reported IC50 values were used in the PTA analysis [9]. The Gelman-Rubin statistic Rhat and effective sample size (ESS) were computed for each parameter distribution, along with trace plots of MCMC chains, density plots of posterior distributions, and posterior predictive checks (PPC), to evaluate model performance. Model development was performed using Stan v2.27.0/Torsten v0.89 with R package cmdstanr v0.61.
Results: Rhat values were close to 1 and ESS values were above 0.5 indicating successful model convergence and reliable estimates of the posterior distribution. Trace plots showed robust mixing of posterior samples, and overlaid density plots from all chains demonstrated convergence to a consistent distribution. The developed model successfully predicted observed acyclovir concentrations within a 2-fold error range, with all observations falling within a 95% prediction interval. Regardless of the virus type, the calculated PTA based on plasma and brainECF concentrations for the standard dosing both fell below 90% when the IC50 value exceeded 1.4 mg/L for plasma and 1.2 mg/L for brainECF, suggesting a potentially suboptimal treatment of the current acyclovir dosing regimen. With a dose of 20 mg/kg, an optimal PTA (>90%) was reached for treating HSV-1, yet remained insufficient (<90%) for treating HSV-2 and VZV.
Conclusions: We successfully developed a CNS PBPK model for acyclovir using the Bayesian methodology. Such an approach allows to integrate information from observed data with prior knowledge to inform model development and address parameter uncertainty. Based on the model simulation, current acyclovir dosing regimen may not suffice for treating herpesvirus-associated CNS infection, and a dose of 20 mg/kg is advised.
References:
[1] Gurgel Assis, M.S., et al., Novel Insights to Enhance Therapeutics With Acyclovir in the Management of Herpes Simplex Encephalitis. J Pharm Sci, 2021. 110(4): p. 1557-1571.
[2] Whitley, R.J., et al., Vidarabine versus acyclovir therapy in herpes simplex encephalitis. N Engl J Med, 1986. 314(3): p. 144-9.
[3] Sköldenberg, B., et al., Acyclovir versus vidarabine in herpes simplex encephalitis. Randomised multicentre study in consecutive Swedish patients. Lancet, 1984. 2(8405): p. 707-11.
[4] Tunkel, A.R., et al., The management of encephalitis: clinical practice guidelines by the Infectious Diseases Society of America. Clin Infect Dis, 2008. 47(3): p. 303-27.
[5] Elmokadem, A., et al., Bayesian PBPK modeling using R/Stan/Torsten and Julia/SciML/Turing.Jl. CPT Pharmacometrics Syst Pharmacol, 2023. 12(3): p. 300-310.
[6] Saleh, M.A.A., et al., Lumbar cerebrospinal fluid-to-brain extracellular fluid surrogacy is context-specific: insights from LeiCNS-PK3.0 simulations. J Pharmacokinet Pharmacodyn, 2021. 48(5): p. 725-741.
[7] Smith, J.P., et al., Pharmacokinetics of acyclovir and its metabolites in cerebrospinal fluid and systemic circulation after administration of high-dose valacyclovir in subjects with normal and impaired renal function. Antimicrob Agents Chemother, 2010. 54(3): p. 1146-51.
[8] Lycke, J., C. Malmeström, and L. Ståhle, Acyclovir levels in serum and cerebrospinal fluid after oral administration of valacyclovir. Antimicrob Agents Chemother, 2003. 47(8): p. 2438-41.
[9] Aoki, F.Y., 45 – Antivirals against Herpes Viruses, in Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases (Eighth Edition), J.E. Bennett, R. Dolin, and M.J. Blaser, Editors. 2015, W.B. Saunders: Philadelphia. p. 546-562.e7.
Reference: PAGE 32 (2024) Abstr 11118 [www.page-meeting.org/?abstract=11118]
Poster: Drug/Disease Modelling - CNS